OECD’s 2026 SBTI Safe Harbour and UAE Zero Rate Incentive

The question to explore, and why most UAE 0% regimes fall away quickly

1. The UAE has multiple “0%” outcomes in its corporate tax architecture, but not all of them share the same logic. For purposes of the SBTI Safe Harbour test, the question is not “is it 0%?” but “is the incentive amount computed in the way the OECD requires”?

2. If we apply that screening lens, most UAE 0% outcomes are immediately vulnerable because they look like income-based preferential regimes: they apply 0% to a category of income (Qualifying Income), typically determined by activity classification, counterparty status, or jurisdictional boundary rules.

3. From the OECD perspective, that is often the hallmark of the kind of preferential regime Pillar Two was designed to neutralize. A regime can be perfectly legitimate under domestic law and still be “income-based” in the Pillar Two sense.

4. That is why, when we ask whether any UAE 0% incentives plausibly qualify as QTIs, only one candidate looks structurally plausible: 0% on Qualifying Income generated by Qualifying Intellectual Property (QIP), calculated using a nexus-style fraction.

5. The reason is simple: unlike most activity-based 0% outcomes, the QIP mechanism is built around a formula that explicitly uses R&D expenditure composition to determine the portion of IP income eligible for 0%. That makes it at least prima facie compatible with the OECD’s insistence on expenditure-based computation.

6. So, the further research narrows to one core inquiry: Is a nexus-ratio IP regime “calculated based on expenditures incurred” in the OECD sense, or is it ultimately an income-based preferential rate with an expenditure-informed attribution key?

Why manufacturing and processing at 0% in a Free Zone is not a QTI in design terms

7. It is tempting to think that “manufacturing” should be a natural fit for “substance-based” treatment because it involves obvious real activity: factories, employees, machinery, and supply chains.

8. But the OECD definition is not a “substance vibes” test. It is a computation test.

9. A 0% rate for manufacturing and processing in a Free Zone typically works by saying: “income from a listed qualifying activity is taxed at 0%”. That is still a preferential rate applied to income, computed from profit, with eligibility driven by activity classification.

10. The OECD’s QTI definition, however, is looking for something different:

• a benefit calculated from expenditure (for example, a credit equal to a percentage of qualifying costs), or
• a benefit calculated from production units (for example, a per-unit subsidy).

11. A manufacturing incentive can qualify in OECD terms if it is genuinely production-based (units produced), or expenditure-based (creditor deduction tied to spend). A manufacturing regime that instead gives “0% on profits from manufacturing” is, in Pillar Two design logic, still an income-based incentive, even if the underlying activity is deeply substantive.

12. So manufacturing/processing being “real activity” does not, by itself, convert an income-based preferential rate into a QTI.

13. That is why the IP regime is the only plausible candidate: it is the only one where the portion of income receiving 0% is determined by a systematic link to expenditure categories.

The IP 0% regime: why it is plausible, and why it is still contentious

The “plausible” reading: nexus looks like expenditure-linked eligible income

14. At first glance, a nexus-style approach appears to fit the OECD’s described architecture for a preferential rate where eligible income is determined using expenditure linkage.

15. In simplified form (ignoring uplift and technicalities), the idea is:

• You compute a nexus fraction that reflects the proportion of “qualifying” R&D over total R&D (or total relevant expenditure base, depending on the regime’s exact definitions).
• You apply that fraction to overall IP income to determine the portion that receives the preferential rate (here, 0%).
• The value of the benefit is then the preferentially taxed income multiplied by the rate differential (statutory rate minus 0%).

This structure is, in a broad economic sense, the “same shape” as an incentive where eligible income is computed using an expenditure-derived percentage.

16. That reading is also supported by the OECD’s own Pillar Two logic, which repeatedly uses the idea of income “attributable to eligible expenditure” as the relevant bridge between expenditure data and income outcomes in safe harbour mechanics.

The alternative view: nexus ratio is not “income determined as a percentage of expenditure” in the relevant sense

17. The competing interpretation, central to the debate, can be stated as follows: Qualifying Income is determined as a portion of Overall Income, not as an amount that is itself “calculated based on a portion of expenditures incurred.” Expenditure is used only to determine an intermediate allocation key. Once that key is computed, applying it to income no longer interacts with the quantum of expenditure. The nexus ratio is therefore best understood as a localization (provenance) measure rather than a spend-based computation base.

18. This is not a semantic point; it is a structural one. A conventional expenditure-based incentive is scale-sensitive: if you double qualifying spend, you ordinarily increase the amount of incentive (credit, super-deduction, allowance). The incentive is computed “off” the spend base.

19. A nexus fraction is different. It is essentially a composition metric: it measures the relationship between “qualifying” and “non-qualifying” inputs, and uses that relationship to justify allocating a share of the income stream into a preferential bucket. It does not, by its nature, tie the magnitude of the benefit to the magnitude of the spending base.

20. The simplest way to see this is with the scale-invariance example:

• Scenario A: qualifying R&D is AED 9 out of total AED 10. Here, nexus ratio equals 90%.
• Scenario B: qualifying R&D is AED 90 million out of total AED 100 million. Nexus ratio equals 90%, as well.

If Qualifying IP income (overall) is AED 1 million in both scenarios, then AED 900,000 enjoys the 0% rate in both scenarios.

The ratio has not “rewarded” the higher spender. It has rewarded the same composition – the same provenance mix. That is why one can argue, with some force, that the nexus formula is not an “expenditure-based incentive” in the ordinary fiscal design sense, but an income-based preferential regime with a provenance gate.

Why this matters for the OECD “benefit exceeds expenditure” exclusion

21. This alternative view becomes especially powerful once you confront the OECD’s exclusion sentence: “An expenditure-based tax incentive is not qualified if the value of the tax benefit from the incentive exceeds the amount of expenditure incurred”. That exclusion fits comfortably with the classic universe of expenditure credits and super-deductions: those incentives are meant to be tethered to spend. If a regime claims to be “expenditure-based” but can produce benefits wildly in excess of expenditure, it begins to behave like an income-based regime masquerading as a spend incentive.

22. But for IP, “benefit exceeds expenditure” is not necessarily suspicious. It is often exactly what happens when IP is successful. High returns on R&D are the economic point of the exercise. If one mechanically disqualifies an IP nexus regime whenever the tax value of the benefit exceeds the R&D cost base, one risks turning the exclusion into a test of commercial success rather than a test of incentive design.

23. This is why the “benefit exceeds expenditure” sentence is so central. It can be read in at least two ways:

• Literal disqualifier reading: if, in a given year (or relevant measurement period), the tax value of the benefit exceeds the expenditure base used to compute eligibility, the incentive is not a QTI.
• Design-targeted reading: the exclusion is aimed at “true expenditure credits” (and similar instruments) to prevent disguised income-based incentives. It should not be applied in a way that makes nexus attribution regimes fail simply because output is high.

The OECD text itself does not spell out a neat reconciliation, which is why the debate is real rather than artificial.

Practical lens: does the rule apply regime-wide, or taxpayer-by-taxpayer?

24. A separate implementation question is how the “benefit exceeds expenditure” exclusion should apply in practice.

25. A strong argument can be made that it is inherently a case-by-case test rather than a regime-wide condemnation, because it requires measurement of:

• the value of the tax benefit, and
• the expenditure incurred (and, critically, which expenditure base is the relevant comparator).

26. Those are not abstract properties of a statute. They are realised in the facts of particular taxpayers and projects. It would therefore be conceptually odd to disqualify an entire regime for all taxpayers, including those whose benefit has never exceeded expenditure, based on the possibility that it might do so for others.

27. This is especially intuitive for IP: the same statutory structure could produce a low benefit in early years (when income is low and spend is high), and a very high benefit later (when income rises and incremental spend falls). A regime-wide disqualification would therefore impose a “worst-case taxpayer” logic on everyone, which does not naturally follow from a test framed in terms of “value of the tax benefit” and “expenditure incurred.”

28. However, one must also acknowledge an uncomfortable implication: if the OECD intends the exclusion to function as a design filter, it may be attracted to a broader view, effectively saying that a regime whose structure can predictably generate benefits exceeding expenditure is not what the QTI category is meant to capture. That pushes the analysis back into the design classification debate discussed above.

Where this leaves the UAE QIP 0% regime

The conservative conclusion

29. A cautious, “audit-proof” conclusion is that most UAE 0% regimes are not QTIs because they are income-based preferential regimes rather than incentives computed from expenditure or production units. On that conservative view, the QIP nexus regime is closer, but still faces real difficulty because (i) the nexus fraction behaves like an attribution key rather than a spend base, and (ii) the “benefit exceeds expenditure” sentence (if applied literally) can disqualify many commercially successful IP cases.

The protective conclusion (the best available defense)

30. A defensible counter-position is that the nexus architecture is precisely an attempt to identify the portion of income that is attributable to qualifying expenditure, and that the OECD’s own safe harbour mechanics are built around that “attribution to expenditure” concept. On that view, the QIP regime remains a plausible candidate for QTI status, especially in cases where the taxpayer can demonstrate that the tax value of the benefit is not disproportionate to the relevant expenditure base when measured appropriately.

31. The strength of the protective conclusion is not that it dissolves the ambiguity. It is that it frames the regime as substance-attribution by design, rather than a profit box.

Conclusion

32. If the OECD’s QTI definition is read narrowly, requiring the incentive amount to be computed directly from spend, and applying the “benefit exceeds expenditure” sentence as a literal disqualifier, then UAE’s 0% Free Zone regimes are out, and even the QIP nexus regime is at significant risk of falling outside QTI status in many real-world fact patterns.

33. If, instead, the definition is read in a more purposive way, accepting that eligible income can be computed via an expenditure-based attribution method, and treating the exclusion as aimed primarily at “disguised income-based credits” rather than successful IP returns, then the QIP 0% regime is the only plausible UAE 0% candidate for QSBTI treatment, while other 0% regimes remain structurally income-based and therefore non-qualifying.

34. Either way, the analysis strongly suggests that “0%” alone is not the relevant feature. The decisive question is whether the regime is computationally tethered to expenditure or production in the OECD sense, or whether it is a preferential rate on income with substance-colored eligibility constraints.

Disclaimer

Pursuant to the MoF’s press-release issued on 19 May 2023 “a number of posts circulating on social media and other platforms that are issued by private parties, contain inaccurate and unreliable interpretations and analyses of Corporate Tax”.

The Ministry issued a reminder that official sources of information on Federal Taxes in the UAE are the MoF and FTA only. Therefore, analyses that are not based on official publications by the MoF and FTA, or have not been commissioned by them, are unreliable and may contain misleading interpretations of the law. See the full press release here.

You should factor this in when dealing with this article as well. It is not commissioned by the MoF or FTA. The interpretation, conclusions, proposals, surmises, guesswork, etc., it comprises have the status of the author’s opinion only. Furthermore, it is not legal or tax advice. Like any human job, it may contain inaccuracies and mistakes that I have tried my best to avoid. If you find any inaccuracies or errors, please let me know so that I can make corrections.